Why errors?
Physics as a natural science subject has its foundation rooted in measurements. For example, in his famous experiment at Pisa Leaning Tower, Galileo showed that the time taken for two spheres of different masses are the same regardless of their mass differences. This experiment disproved Aristotle’s theory of gravity (which states that objects fall at speed proportional to their mass). This clearly showed that experiment is the ultimate test for any scientific theory. However, before any conclusion can be drawn from an experiment, the accuracy and precision of the experimental data needs to be verified. Try this falling ball experiment.
#DEFINITION#
Systematic errors results in all readings of measurements being always smaller or always larger than the true value by a fixed amount (RJC 2017).
Tee’s Physics:
Systematic error indicates fundamental fault in experiment which lead to wrong conclusions of the physics.
Normally, systematic errors occur in the following scenarios:
- Measuring equipment is not calibrated properly with zero error being the common one (e.g. non-zero reading on digital vernier caliper). Note: Some calibration-related error could be more complicated than zero error.
- Experiment method is wrong which means certain physics effect or interaction is not captured in mathematical modelling. (e.g. free fall experiment in viscous liquid where the liquid resistance is not represented in the free fall equation
F=mg).
So what to do? To remove systematic error, one needs to seriously identify and eliminate the source of error. Otherwise, the interpretation of results could lead to conclusion of wrong physics. For examples,
- Performing free fall experiment in viscous liquid is erroneous because the liquid presents a resistance force to the falling object. Solution: do the experiment in vacuum which completely eliminate resistance to free fall.
- Always zero any measuring equipment to be used.
DEFINITION#
Random errors result in readings or measurements scattered about a mean value. (RJC 2017).
Tee’s Physics:
Random errors indicate limitations in measuring capability which lead to uncertainties in the quantitative results.
Examples of random errors include:
- Measuring equipment is not calibrated properly with zero error being the common one (e.g. non-zero reading on digital vernier caliper). Note: Some calibration-related error could be more complicated than zero error.
- Experiment method is wrong which means certain physics effect or interaction is not captured in mathematical modelling. (e.g. free fall experiment in viscous liquid where the liquid resistance is not represented in the free fall equation
F=mg).
Random errors can be reduced by
- Repeat measurements for many times to find the mean value.
- Apply the best fit line to curve fit all data points.
Tee's Physics 